CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

bystander populations. Following this analysis, 50 genes were selected based on their differential expression in HIV-1-infected macrophages. Within those genes, we found that the expression of the folate cycle enzyme Gamma- Glutamyl Hydrolase (GGH) was upregulated in infected cells early after infection. We suggest that, in primary human macrophages, folate derivatives like tetrahydrofolate (THF) may play a protective role against infection by HIV-1. Methods: Monocyte-derived macrophages (MDMs) were transfected with small interfering RNAs (siRNAs) targeting folate cycle enzymes GGH, Folylpolyglutamate Synthase (FPGS) and Methylenetetrahydrofolate Reductase (MTHFR) using lipofectamine RNAiMAX™ 3 days prior to infection with HIV-1. In other experiments, MDMs were exposed to increasing concentrations of Raltitrexed (RTX), a specific inhibitor of Thymidylate Synthase (TS), to assess the effect of TS inhibition during HIV-1 infection. We developed an experimental model based on the identification of productively infected MDMs by a fully replicative molecular clone of R5-tropic HIV-1 expressing all viral genes and a small GPI-anchored reporter gene (murine Heat Stable Antigen; HSA). After infection (20 ng of p24 per 105 cells), percentages of productively infected cells were evaluated by flow cytometry and ELISA targeting the viral capsid (p24) at 3 to 12 days post-infection. Results: Downregulation of GGH, FPGS, and MTHFR gene expression increases initial HIV1 infection in MDMs by 22%, 65% and 51% respectively. Viral production kinetics are similarly enhanced. Also, by targeting TS activity with RTX, we significantly reduce infection. This inhibitory effect can last several days without affecting cell viability. Conclusion: Interplay between HIV-1 and the folate cycle may be a key factor in the susceptibility of MDMs to HIV-1 infection. Downregulation of enzymes that involved in intracellular folate retention increase the percentage of productively infected cells. Folate derivatives like THF may thus to be important cofactors in the restriction of HIV-1 infection in MDMs. Finally, TS could be explored as a target for intervention against HIV-1 infection in macrophages. constitution of a long-lived reservoir inside target cells, such as CD4+ T cells and macrophages. Based on in vitro observations, only a small proportion of macrophages is initially susceptible and could contribute to the establishment of a reservoir early after infection. To better understand molecular mechanisms of macrophages permissiveness, we performed transcriptomic analysis of infected and uninfected bystander cells. Precise mapping of the interactions between HIV-1 and host signaling pathways identified potential novel regulators of HIV-1 replication in macrophages. One of these was a member of the lysyl oxidase family, Lysyl-Oxidase-like-3 (LOXL3), which is known to regulate collagen/ elastin crosslinking. Methods: Our team has developed a fully competent molecular clone of HIV-1 bearing a GPI-anchored reporter gene allowing the identification and separation of productively infected Monocyte-Derived Macrophages (MDMs). Sorting of infected and bystander cells allowed us to compare transcriptomic signatures within these two populations. In so doing, 50 modulated genes were screened for their effects on HIV-1 infection by siRNA-mediated knockdown. Among these, LOXL3 knockdown was found to significantly impact susceptibility to infection. Validation experiments were thus initiated to explore its effects on viral replication and infection of HIV-1 within macrophages and compare it to known susceptibility (CD4) and restriction (SAMHD1) factors. Results: Gene silencing of LOXL3 results in a 2-fold increase in the percentage of MDMs infected by HIV-1 as early as 2-days post infection and in a significant enhancement of viral production. Interestingly, LOXL3 knockdown leads to an upregulation of CD4 expression at both mRNA and protein levels. However, downregulation of LOXL3 gene also increases infection by a single-round VSV-G- pseudotyped HIV-1, which mode of entry is CD4-independent. Conclusion: Although increase in CD4 expression following LOXL3 gene silencing suggests a potential explanation for the observed enhanced susceptibility to infection, VSV-g pseudotyping experiments instead imply a post-entry mechanism of action. Nevertheless, our data suggest that LOXL3 could be a restriction factor for HIV-1 replication in primary human macrophages. Identification of the mechanisms underlying LOXL3-mediated

restriction will bring a better understanding of signaling events controlling HIV-1 replication in macrophages. 180 SYNONYMOUS RECODED ENV GENE INDUCE LETHALITY AND LOSS OF PROTEIN EXPRESSION IN HIV-1 Ana Jordan-Paiz , Maria Nevot, Sandra Franco, Miguel Angel Martinez IrsiCaixa Institute for AIDS Research, Badalona, Spain Background: For the expression of HIV-1 late genes, the early expressed transinducer Rev is needed to transport viral mRNAs to the cytoplasm (e.g. Env mRNA). The genetic code is redundant and codon usage differs among different species. Protein expression is affected by the presence of “rare” codons. We aim here to explore the impact of synonymous codon usage on HIV-1 ENV expression and virus replication capacity. Methods: Six codons, AGG, GAG, CCT, ACT, CTC and GGG of HIV-1 env gene were synonymously changed to CGT, GAA, CCG, ACG, TTA and GGA, respectively (as described by Shin et al, PNAS 2015); consequently 39 mutations were introduced. To obtain viral particles, the different parts of HIV-1 genome were PCR amplified and transfected in MT-4 cells. Viral replication was quantified by measuring HIV‐1 capsid p24 antigen production. WT and recoded env genes were also introduced in an expression vector (pcDNA3.1) and, together with the rev gene, were transfected in 293T cells. Protein expression and Env mRNA production were determined by WB and qPCR, respectively. Results: After transfection, we observed that the synonymously recoded Env gene (recoded-Env) was lethal for the virus; no syncitia and no p24 production were observed after five blind passages in MT-4 cells. WB analysis of Env expression in 293T cells revealed that, in contrast to WT, the recoded-Env mRNA protein was not being translated although there were not significant differences in Env mRNA production. Additional mutants were designed to see which mutations were responsible for this phenotype, without modifying env codon usage (CAI) or codon pair bias (CPB). Newmutants included changes in total number substitutions and/or CpG content (see Table). All newmutant env genes generated replicative viruses, however, they displayed different replicative capacities. Remarkably, synonymous substitutions at the 3’ end of the env gene were affecting, to a greater extent, the virus phenotype. Conclusion: We show here that the HIV-1 replication capacity is affected by the codon usage. Our results also indicate that mutations in the 3’ coding region of Env (gp41) can lead to lethality. Ex vivo expression experiments demonstrated that Env mRNA translation was affected. However, it remains to be elucidated whether the interaction of Env mRNA and Rev is affected. Overall, our results emphasize the relevance of synonymous substitutions in shaping virus phenotype.

Poster Abstracts

179 LOXL3 REGULATES HIV-1 INFECTION IN HUMAN PRIMARY MACROPHAGES Laurent Hany , Michel Ouellet, Michel Tremblay Laval University, Quebec City, QC, Canada Background: One of the suspected reason for HIV persistence is the

181 ENHANCEMENT OF HIV-1 INFECTION BY BUPRENORPHINE

German G. Gornalusse 1 , Lucia Vojtech 1 , Claire Levy 1 , Rogelio M. Valdez 2 , Urvashi Pandey 2 , Julie McElrath 2 , Florian Hladik 1 1 University of Washington, Seattle, WA, USA, 2 Fred Hutchinson Cancer Research Center, Seattle, WA, USA Background: Medication-assisted treatment (MAT) with buprenorphine is now widely prescribed to treat addiction to heroin and other illicit opioids. There is some evidence that illicit opioids enhance HIV-1 replication and accelerate AIDS pathogenesis, but the effect of buprenorphine is unknown. Methods: We obtained peripheral blood mononuclear cells (PBMCs) from healthy volunteers and activated themwith phytohemagglutinin and interleukin 2 for 72 hours in the presence of morphine or buprenorphine. We infected the cells with a panel of replication-competent CCR5-tropic HIV-1 reporter viruses encoding a secreted nanoluciferase gene, and measured infection by luciferase activity in the supernatants over time. We also surveyed opioid receptor expression in PBMC and vaginal leukocytes, as well as Langerhans cells derived from CD34+ hematopoietic stem cells, by qPCR.

64

CROI 2018